LVS+keepalived实现双机热备

昨晚写了关于lvs的文章，今天以实例介绍下LVS+keepalived实现双机热备

服务器IP地址设置表

部署两台或者多台LVS调度器,仅有一台调度器作为主服务器,其他调度器作为备用,当主调度器发生故障时,keepalived会自动将备用服务器升为主服务器

1.lvsmaster端

下载keepalived服务

yum install -y keepalived

源码安装

yum install -y openssl openssl-develwget http://www.keepalived.org/software/keepalived-1.2.12.tar.gztar xvf keepalived-1.2.12.tar.gzcd keepalived-1.2.12./configure --bindir=/usr/bin --sbindir=/usr/sbin --sysconfdir=/etc --mandir=/usr/sharemake && make install

配置keepalived

[[email protected] ~]# vim /etc/keepalived/keepalived.conf ! Configuration File for keepalived

global_defs {   notification_email {        [email protected]   }## 默认的配置文件中,使用第三方smtp服务器,但这在现实中几乎没有意义（需要验证的原因）,我们将其指定为localhost, 将通知信息的发送交给本地sendmail服务处理   notification_email_from [email protected]   smtp_server 127.0.0.1   smtp_connect_timeout 30   router_id LVS_DEVEL}

vrrp_instance VI_1 {    state MASTER         ## 从使用BACKUP    interface eth0          ## 确认对外开放的端口,这里设置错误,将导致不会建立VIP    virtual_router_id 51    priority 100            ## 从要比主低,我设置为80    advert_int 5    mcast_src_ip 172.16.100.2  ##lvs的IP,从IP为172.16.100.3    authentication {        auth_type PASS        auth_pass Golo    }

# notify_master,notify_backup和notify_fault# 分别表示在节点变为master，节点变为backup和节点不可用时执行的脚本    notify_master "/etc/keepalived/master.sh"    notify_backup "/etc/keepalived/backup.sh"    notify_fault "/etc/keepalived/fault.sh"

## 在keepalived的配置文件最前面加入以下代码，定义一个跟踪脚本：(解决脑裂问题) vrrp_script check_local { #定义一个名称为check_local的检查脚本    script "/usr/local/keepalived/bin/check_local.sh" #shell脚本的路径interval 5  #运行间隔weight -15 #优先级变化}

## 再在vrrp_instance配置中加入以下代码使用上面定义的检测脚本：track_script {    check_local}

    virtual_ipaddress {       172.16.100.100     ## VIP    }}virtual_server 172.16.100.100 80 {    delay_loop 6    lb_algo rr    lb_kind DR    persistence_timeout 60    protocol TCP

    real_server 172.16.100.2 80 {        weight 3        TCP_CHECK {        connect_timeout 10        nb_get_retry 3        delay_before_retry 3        connect_port 80}    }    real_server 172.16.100.3 80 {        weight 3        TCP_CHECK {        connect_timeout 10        nb_get_retry 3        delay_before_retry 3        connect_port 80        }     }}

2.lvsbackup端

lvs-backup的修改参照上面的注释部分

开启keepalived服务

/etc/init.d/keepalived start

执行ipvsadm可以看到keepalived中配置已经在ipvsadm中调用

[[email protected] ~]# ipvsadmIP Virtual Server version 1.2.1 (size=4096)Prot LocalAddress:Port Scheduler Flags  -> RemoteAddress:Port           Forward Weight ActiveConn InActConnTCP  172.16.100.100:http rr persistent 50  -> web1:http                    Route   3      0          0           -> web2:http                    Route   3      0          0      

3.web端

web端配置于前篇lvs文章中一样,这里写成脚本形式开机自启动

在/etc/init.d/下web1 172.16.100.4和web2 172.16.100.5上的脚本lvs

内容如下：

#add for chkconfig#chkconfig: 2345 70 30  #description: lvs‘s script#processname: lvs#!/bin/bashVIP=172.16.100.100source /etc/rc.d/init.d/functionscase "$1" instart)       ifconfig lo:0 $VIP netmask 255.255.255.255 broadcast $VIP       /sbin/route add -host $VIP dev lo:0       echo "1" >/proc/sys/net/ipv4/conf/lo/arp_ignore       echo "2" >/proc/sys/net/ipv4/conf/lo/arp_announce       echo "1" >/proc/sys/net/ipv4/conf/all/arp_ignore       echo "2" >/proc/sys/net/ipv4/conf/all/arp_announce       echo "RealServer Start OK"       ;;stop)       ifconfig lo:0 down       route del $VIP >/dev/null 2>&1       echo "0" >/proc/sys/net/ipv4/conf/lo/arp_ignore       echo "0" >/proc/sys/net/ipv4/conf/lo/arp_announce       echo "0" >/proc/sys/net/ipv4/conf/all/arp_ignore       echo "0" >/proc/sys/net/ipv4/conf/all/arp_announce       echo "RealServerStoped"       ;;       *)       echo "Usage: $0 {start|stop}"       exit 1esacexit 0

为脚本lvs添加权限,并启动

chkconfig lvs onservice lvs start

安装并启动nginx，这里不多做介绍

4.结果验证

在lvs服务器上查看系统日志,服务启动成功

[[email protected] ~]# tail -f -n 100 /var/log/messagesAug 13 16:03:25 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: IPVS: Service already existsAug 13 16:03:25 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: IPVS: Destination already existsAug 13 16:03:25 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: IPVS: Destination already existsAug 13 16:03:25 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: Using LinkWatch kernel netlink reflector...Aug 13 16:03:25 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: Activating healthchecker for service [172.16.100.4]:80Aug 13 16:03:25 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: Activating healthchecker for service [172.16.100.5]:80Aug 13 16:03:34 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: Remote SMTP server [127.0.0.1]:25 connected.Aug 13 16:03:34 sz-mycar-lvssla1 Keepalived_healthcheckers[20193]: SMTP alert successfully sent.

查看lvsmaster,可看到VIP在master上

[[email protected] ~]# ip a li eth22: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000    link/ether 52:54:52:4f:2a:de brd ff:ff:ff:ff:ff:ff    inet 172.16.100.2/24 brd 172.16.100.255 scope global eth2    inet 172.16.100.100/32 scope global eth2    inet6 fe80::5054:52ff:fe4f:2ade/64 scope link        valid_lft forever preferred_lft forever

将lvsmaster上的keepalived服务关闭,可发现VIP已经切换到lvsbackup

[[email protected] ~]# ip a li eth22: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000    link/ether 52:54:2b:55:07:d9 brd ff:ff:ff:ff:ff:ff    inet 172.16.100.3/24 brd 172.16.100.255 scope global eth2    inet 172.16.100.100/32 scope global eth2    inet6 fe80::5054:2bff:fe55:7d9/64 scope link        valid_lft forever preferred_lft forever

主从切换成功,且服务照常运行